AI consultant in Guadalajara for businesses
We turn your company's processes into AI systems that work with your data, your software and your team. Consulting, implementation, integration and measurement by the same team from start to finish, at your offices.
Guadalajara, Jalisco
What we do, in plain terms
WITS is an artificial intelligence and software development consultancy based in Guadalajara. We work with mid-sized companies and SMBs in the region and with clients across the rest of Mexico.
We don't sell licenses or a generic chatbot. We step into your operation, identify the processes where AI has a real return and build them: AI agents, process automation, integrations with your CRM, ERP and WhatsApp, and custom software when it's needed.
The difference from a global firm is simple: the team that runs the assessment is the same team that builds, we charge like a local firm, and the result is measured in systems running, not in slide decks.
We build it, not just recommend it
AI chat for automation
Web and mobile AI chat that generates automation projects and handles product search, connected to an ecommerce store.
AI-powered asset management
Assets and liabilities system connected to Telegram and ChatGPT.
State investment CRM
Investment project management platform for the Government of Jalisco.
What can AI do inside your company?
The answer is almost never one more tool. It's a system that reads, decides and acts on what you already have. These are the areas where we usually start.
Customer service
Replies over WhatsApp and email with context from your system, ticket classification and routing, order tracking, and handoff to a person when needed.
Sales
Lead qualification, automatic first contact and follow-up, and proposal drafts that a salesperson reviews before sending.
Operations
Data extraction from invoices, orders and PDFs, automatic entry into your ERP, and alerts when an order or delivery leaves the expected flow.
Administration and finance
Bank and vendor reconciliations, invoice checks against purchase orders, and the weekly report someone builds today by copying from three systems.
Engineering and IT
Log and alert analysis, internal assistants over your documentation and your code, and automation of support and deployment tasks.
Leadership
Dashboards that combine several systems, and agents that query your ERP, CRM and databases to answer business questions in natural language.
Problems we solve
- We get AI proposals every week and have no one to review them with who isn't trying to sell us something.
- Leadership wants to do something with AI and nobody in the company knows what to ask for or who to ask.
- We have data in the ERP, in the CRM and in a pile of spreadsheets, and nobody knows whether any of it is usable for AI.
- Quotes, orders and reports get entered two and three times.
- A vendor gave us an impressive demo and in production it didn't work with our data.
- We can't put sensitive company information into just any tool.
How we work
Three stages, each with a deliverable and the option to stop if it doesn't make sense to continue.
- 1
Assessment
Sessions with your team, at your offices. We map processes, systems and data, and come out with a prioritized list of use cases, each with its estimated cost and the expected impact defined with you. If the honest answer is "AI doesn't make sense here yet", we say so.
- 2
Implementation
We build the first use case connected to your real systems, with permissions and human approvals where the process calls for them. Short cycles, a demo every week, metrics defined before we start.
- 3
Ongoing support
We measure against the agreed metrics, adjust and extend to the next processes. Your team learns to operate the system and to ask for what comes next.
No made-up percentages: the metrics are defined with you before we build and measured afterward.
What you get from the assessment
A working document that comes out of sessions with your team, not a questionnaire. It helps you decide, whether you build with us or with someone else.
- A map of the processes reviewed: who does what, with which system and where it gets stuck.
- Automation opportunities, with and without AI, separated from the ones solved with a rule or a form.
- Prioritized AI use cases, with the scope of each one written on a single page.
- Systems and data involved in each case: what gets integrated, what's missing and what needs cleanup.
- Effort and cost estimate per use case, so the number exists before you decide.
- Expected impact defined with you: which metric moves and how it will be measured.
- Recommended priority: where to start and what to leave for later.
- Our written recommendation: do it, don't do it or wait, and why.
We're in the metro area, not on a video call
For the first meeting we come to your company, in any municipality of the metropolitan area. We'd rather see the process where it happens before giving an opinion on it.
We have worked with the State Government (SEDECO) and with industrial automation, fintech and software companies. We know how family-owned and mid-sized businesses here operate: legacy systems, a lot of WhatsApp and little tolerance for projects that never finish.
WITS is a good fit if…
Not every company needs an AI consultant. If several of these sound like you, a conversation is worth it.
- You have a list of AI ideas and need someone to sort them, price them and rule out the ones that don't pay off.
- You're about to invest in licenses, a vendor or new hires, and want an informed decision first.
- You don't have an in-house data or AI team and don't want to hire one before knowing what to build.
- Your ERP, your CRM or your own platform won't be replaced for this project: whatever gets built has to live alongside them.
- You can assign a business-side owner to the project, even if they're not from IT, to validate what gets built.
- You need what gets built to stay on your infrastructure and under your control, not on a third party's platform.
What people ask before hiring us
How many in-person sessions does the assessment take?
Usually two or three at your offices: one with leadership for the objective, and one or two with the people who run the process to review systems and data. We do the rest of the work and come back to present the document.
Do you work with small companies or only large ones?
We work with mid-sized companies and with SMBs that already have a specific process that hurts. What matters is not size but having a repetitive process, with volume and data, where AI changes something measurable.
How much does the consulting cost?
The first visit is free. The assessment is a short, fixed-price project whose price you know before we start. Implementation is quoted per use case once the scope is defined in the assessment.
What about the security of our data?
Before touching any data we define with you which data can leave the company, to which provider and under what agreement. When the case requires it, the model runs in a private deployment. Nothing we build stays on a platform of ours: code, access and data remain in your company's name.
What technology do you work with?
Language models from Anthropic (Claude) and OpenAI, agents with tool use, API integrations with CRM, ERP, WhatsApp Business and databases, and custom development when the process requires it. The technology is chosen after the problem, not before.
How long until we see a result?
The assessment delivers the prioritized list in weeks. The first use case in production usually takes from a few weeks to a few months, depending on how much needs to be integrated with your systems. We measure before and after with the metrics we agree on.
Let's look at the process where it happens
Tell us which process you want to improve. We come to your company, review it with the people who run it and tell you whether it makes sense to tackle it with AI, automation or software, and what it would take to put a number on it. No cost, no commitment.